Study reference

Study characteristics

Patient characteristics

Intervention (I)

Comparison / control (C)

Follow-up

Outcome measures and effect size

Comments

Matchett,

2022

Type of study:

Single-center RCT

Setting and country:

high-volume medical center, USA

Funding and conflicts of interest:

Internal funding by Dept. of Anesthesiology & Pain Management.

One author receives compensation for serving on the Medical Advisory Board for the industry.

Inclusion criteria:

adult patient 18

years and older in need of emergency endotracheal intubation

Exclusion criteria:

children (< 18

years), women known to be pregnant, patients previously enrolled in the trial, patients requiring endotracheal

intubation without sedative medication such as those in cardiac arrest, those neurologically obtunded, or those

requiring awake intubation, patients

with a known allergy to etomidate or ketamine

N total at baseline:

Intervention: 400

Control: 401

Important prognostic factors²:

age ± SD:

I: 55.8 ± 15.2

C: 55.4 ± 16

Sex:

I: 61.4% M

C: 62% M

Groups comparable at baseline?

Yes

Describe intervention (treatment/procedure/test):

Etomidate 0.2–0.3 mg/kg IV

Describe control (treatment/procedure/test):

ketamine 1–2 mg/kg IV

Length of follow-up:

7, 28 days

Loss-to-follow-up:

Intervention:

N (%): 4 (1%)

Reasons: withdrawals (n=4)

Control:

N (%): 6 (1.5%)

Reasons: withdrawals (n=6)

Incomplete outcome data:

The number of attempts to intubate is missing for five patients in the etomidate group and six patients in the ketamine group, all of whom were successfully intubated using non-surgical techniques according to chart documentation

Outcome measures and effect size (include 95%CI and p-value if available):

Mortality 7 days, n(%)

I: 90/400 (22.5)

C: 59/401 (14.7)

RR 1.53 (1.14 to 2.06)

Mortality 28 days, n(%)

I: 142/396 (35.9)

C: 131/395 (33.2)

First pass success, n(%)

I: 357/396 (91.3)

C: 355/395 (91.3)

Post-induction CPR, n(%)

I: 13/396 (3.8)

C: 18/395 (5.1)

Author’s conclusion:

While the primary outcome of Day 7 survival was greater

in patients randomized to ketamine, there was no significant difference in survival by Day 28.

Punt, 2014

Type of study:

Cluster-randomized trial

Setting and country:

ICU, large teaching hospital, Netherlands

Funding and conflicts of interest:

No information

Inclusion criteria:

critically ill adult patients, who were intubated in one of the intensive care units, either shortly after admission or during their period of stay

Exclusion criteria:

patients who had received etomidate less than 72 hours before

N total at baseline:

Intervention: 161

Control: 140

Important prognostic factors²:

age ± SD:

I: 66 ± 14

C: 67 ± 13

Sex:

I: 59% M

C: 63.7% M

Groups comparable at baseline?

Yes

Describe intervention (treatment/procedure/test):

Etomidate 0.2–0.3 mg/kg

Describe control (treatment/procedure/test):

S-Ketamine 0.5 mg/kg and midazolam 2.5 mg.

Length of follow-up:

28 days

Loss-to-follow-up:

Intervention: not reported, not assumed

Control: not reported, not assumed

Incomplete outcome data:

Intervention: not reported, not assumed

Control: not reported, not assumed

Outcome measures and effect size (include 95%CI and p-value if available):

Mortality 28 days, n(%)

I: 61/161 (37.9)

C: 54/140 (38.6)

RR 0.98 (0.74 to 1.31)

Author’s conclusion:

The present study showed that the 28-day mortality after a single dose of etomidate or S-ketamine, administered to patients for tracheal intubation while on the intensive care did not differ.

Smischney, 2019

Type of study:

Randomized clinical trial

Setting and country:

ICU of a tertiary academic medical center, USA

Funding and conflicts of interest:

Funding received was institutional.

First and second author have a provisional patent filed for the ketamine-propofol admixture.

Inclusion criteria:

Adult patients (≥18 years), admitted to one of three ICUs (medical,

surgical, and oncologic/transplant ICU totaling 65 beds) in a tertiary medical center, requiring emergent intubation from July

2014 to October 2017

Exclusion criteria:

(1) intubation for cardiac arrest; (2) intracranial pathology with known/

suspected elevation in intracranial pressure; (3) chronic opioid dependence defined by methadone, buprenorphine, buprenorphinenaloxone, or naltrexone use as an outpatient; (4) bipolar and/or schizophrenia disorder; (5) egg allergies; (6) contraindications to fentanyl, midazolam, ketamine, propofol, or etomidate; (7) any weight greater than 140 kg or less than 30 kg; (8) previously enrolled in the study; (9) continuous infusions of propofol, midazolam, lorazepam, fentanyl, or dexmedetomidine in the

previous 24 hours; (10) procedural intubation; (11) incarceration; (12) any woman of child-bearing age, between 18 years and 50 years without a pregnancy test.

N total at baseline:

Intervention: 76

Control: 84

Important prognostic factors²:

age ± SD:

I: 60.0 ± 18.3

C: 62.1 ± 17.2

Sex:

I: 52% M

C: 61% M

Groups comparable at baseline?

Yes

Describe intervention (treatment/procedure/test):

An induction intravenous dose of 0.15 mg/kg

(actual body weight) of etomidate was used.

A second dose of 0.15 mg/kg was available as a rescue.

In both arms, a standard 50 μg intravenous bolus dose of fentanyl was administered to blunt the hemodynamic response.

Describe control (treatment/procedure/test):

Ketamine-propofol admixture (KPA):

An initial dose of 1 mg/kg total of KPA (0.5 mg/kg of ketamine and propofol

each) was the standard intravenous induction dose.

A second dose of 1 mg/kg KPA (0.5 mg/kg of ketamine

and propofol each) was available as a rescue.

Length of follow-up:

Hospital discharge

Loss-to-follow-up:

Intervention:

N = 3 (3.9%)

Reasons: refused data-analysis (n=3)

Control:

N = 5 (6.0%)

Reasons: refused data-analysis (n=5)

Incomplete outcome data:

Intervention: not reported, not assumed

Control: not reported, not assumed

Outcome measures and effect size (include 95%CI and p-value if available):

Mortality hospital, n(%)

I: 26/76  (34.2)

C: 25/84 (29.8)

RR 1.15 (0.73 to 1.81)

Vomiting/aspiration*:

I: 3/73 (4%)

C: 2/79 (3%)

RR 1.62 (0.28 to 9.44)

*study states the events are probably not related to the study drug

Cardiac arrest*

I: 0/73

C: 2/79 (3%)

RR 0.22 (0.01 to 4.43)

*study states that probably 1 event in the control group was related to the study drug

Hypotension (MAP <40 mm Hg)*

I: 4/73 (5%)

C: 3/79 (4%)

RR 1.44 (0.33 to 6.23)

*study states that probably 1 event in each group was related to the study drug

Author’s conclusion:

KPA was not superior to a reduced dose of etomidate in terms of hemodynamic profile and new-onset vasopressor need after emergent intubation in critically ill patients. There were no differences in frequency of delirium or intubation difficulty. KPA appears to be a safe alternative induction agent compared with reduced dose etomidate and should be considered whenever adrenal insufficiency is a concern.

Jabre, 2009

Type of study:

randomised, controlled, single-blind trial

Setting and country:

12 emergency medical services or emergency departments and 65 ICUs in France

Funding and conflicts of interest:

Funding by the French Ministry of Health (no role in study design, data collection, data analysis, data interpretation, or writing the report). Nothing else to declare.

Inclusion criteria:

Patients who were 18 years or older and who needed sedation for emergency intubation

Exclusion criteria:

cardiac arrest; contraindications to succinylcholine, ketamine, or etomidate; or known pregnancy

N total at baseline:

Intervention: 328

Control: 327

Important prognostic factors²:

age ± SD:

I: 57 ±  18

C: 59 ±  19

Sex:

I: 63% M

C: 57% M

Groups comparable at baseline?

Yes

Describe intervention (treatment/procedure/test):

Etomidate 0.3 mg/kg

Both groups:

Succinylcholine was given immediately after the sedative as a 1 mg/kg intravenous bolus. After confirmation of intubation and tube placement, continuous sedation was

initiated by use of a standardised protocol with midazolam (0·1 mg/kg/h) combined with fentanyl (2–5 μg/kg/h) or sufentanil (0·2–0·5 μg/kg/h).

Describe control (treatment/procedure/test):

Ketamine 2 mg/kg

Length of follow-up:

28 days

Loss-to-follow-up:

Intervention:

N = 94 (28.7%)

Reasons: withdrew consent (n=1), missing data (n=2) discharged from ICU before 3 days (n=79), died before reaching hospital (n=12)

Control:

N 92 (28.1%)

Reasons: missing data (n=2) discharged from ICU before 3 days (n=75), died before reaching hospital (n=15)

Incomplete outcome data:

Intervention:

NR

Control:

NR

Outcome measures and effect size (include 95%CI and p-value if available):

Mortality 28 days, n

I: 93/328

C: 87/327

MD 1.07 (0.83 to 1.37)

Cardiac arrest during intubation, n

I: 7/328 (3%)

C: 4/327 (2%)

RR 1.74 (0.52 to 5.90)

Srivilaithon, 2023

Type of study:

randomized, controlled, single-blind

trial

Setting and country:

Emergency Medicine Research Group at Tammasat University Hospital (TUH) in Pathum Tani, Thailand

Funding and conflicts of interest:

Academic funding, no competing interests

Inclusion criteria:

Patients presenting to the ED with suspected sepsis who were 18  years or older and then needed an induction agent for emergency intubation in the ED

Exclusion criteria:

1) cardiac arrest before intubation, 2) presence of a do-not-resuscitate order, 3) known or suspected adrenal insufficiency, 4) severe hypertension (blood pressure before randomization > 180/110 mmHg), and 5) suspected or evidenced increased intracranial pressure.

N total at baseline:

Intervention: 130

Control: 130

Important prognostic factors²:

age ± SD:

I: 73.2 (±12.6)

C: 70.5 (±14.9)

Sex:

I: 59.2% M

C: 58.5% M

Groups comparable at baseline?

Yes

Describe intervention (treatment/procedure/test):

single-dose induction

agent consisting of either etomidate (Lipuro, B. Braun Melsungen, Germany) administered as a 0.2–0.3 mg/kg intravenous bolus

Describe control (treatment/procedure/test):

ketamine (Ketalar, PAR Pharmaceutical, Ireland) administered as a 1–2 mg/kg intravenous bolus

Length of follow-up:

28 d

Loss-to-follow-up:

Intervention:

N = 0 (0%)

Reasons (describe)

Control:

N = 0 (0%)

Reasons (describe)

Incomplete outcome data:

None

Outcome measures and effect size (include 95%CI and p-value if available):

24h survival

I: 119/130

C: 125/130

7d survival

I: 114/130

C: 114/130

28d survival

I: 105/130

C: 95/130

Cardiac arrest

I: 2/130

C: 2/130

Post-intubation hypotension

I: 15/130

C: 14/130

Intubation successful in the first attempt

I: 116/130

C: 114/130

Author’s conclusion:

In patients with clinically suspected sepsis who needed emergency intubation in the ED, there was no difference

in early and 28-day survival rates between etomidate and ketamine. However, etomidate was associated with higher risks of early vasopressor use after intubation.

Tekwani, 2010

Type of study:

prospective, double-blind, randomized study

Setting and country:

a large, tertiary care suburban hospital, USA

Funding and conflicts of interest:

Financial support by Emergency Medicine Foundation

Resident Research Grant.

Inclusion criteria:

Older than 18 years, were intubated in the ED, and had a suspected

infectious cause for their illness

Exclusion criteria:

younger than 18 years, pregnancy, cardiopulmonary arrest before arrival in the ED, or a do-not resuscitate status

N total at baseline:

Intervention: 63

Control:59

Important prognostic factors²:

For example

age (IQR):

I: 70 (60 to 83)

C: 73 (60 to 80)

Sex:

I: 52% M

C: 61% M

Groups comparable at baseline?

Yes

Describe intervention (treatment/procedure/test):

Etomidate 0.3 mg/kg IV before RSI

Describe  control (treatment/procedure/test):

Midazolam 0.1 mg/kg before RSI

Length of follow-up:

Hospital stay

Loss-to-follow-up:

Intervention:

N = 2 (3.2%)

Reasons (describe): transfer to outside institutions.

Control:

N = 0 (0%)

Reasons (describe)

Incomplete outcome data:

Not reported, not assumed

Outcome measures and effect size (include 95%CI and p-value if available):

Hospital mortality

I: 26/63

C: 21/59

Low SBP <90 mm Hg

I: 16/63

C: 15/59

Author’s conclusion:

In conclusion, for patients who had suspected sepsis and

were intubated in the ED, we found no significant difference in

hospital length of stay between patients who received etomidate

as an induction agent and those who received midazolam.

Ali, 2021

Type of study:

randomised double-blinded controlled trial.

Setting and country:

surgical

theatre at the emergency department of Cairo University

Hospital, Egypt

Funding and conflicts of interest:

Self-funded, no conflicts of interest

Inclusion criteria:

age between 18 and 65

years; experiencing septic shock; scheduled for emergency

surgery to control the source of sepsis

Exclusion criteria:

burns, traumatic brain

injury or history of any cerebrovascular disorders.

N total at baseline:

Intervention: 23

Control: 23

Important prognostic factors²:

age (IQR):

I: 44 (40-60)

C: 56 (46-63)

Sex:

I: 11% M

C: 8% M

Groups comparable at baseline?

Yes

Describe intervention (treatment/procedure/test):

Before induction: ketamine 1 mg/kg IV

plus midazolam (0.05 mg/kg) and succinyl choline (1 mg/kg).

Describe control (treatment/procedure/test):

Before induction: fentanyl 2.5 mcg/kg IV

plus midazolam (0.05 mg/kg) and succinyl choline (1 mg/kg).

Length of follow-up:

10 minutes

Loss-to-follow-up:

Intervention:

N = 0 (0%)

Reasons (describe)

Control:

N = 4 (17.4%)

Reasons (describe): poor cardiometry (n=2), incomplete data (n=2)

Incomplete outcome data:

See above

Outcome measures and effect size (include 95%CI and p-value if available):

Post-induction hypotension, N (MAP ≤ 80% the baseline value during the period from induction of anaesthesia until 10 min)

I: 11/23

C: 16/19

Author’s conclusion:

In conclusion, the ketamine-based regimen provided a

better hemodynamic profile compared to the fentanyl-based regimen for rapid-sequence intubation in patients with septic shock undergoing emergency surgery

Study reference

(first author, publication year)

Was the allocation sequence adequately generated?

Definitely yes

Probably yes

Probably no

Definitely no

Was the allocation adequately concealed?

Definitely yes

Probably yes

Probably no

Definitely no

Blinding: Was knowledge of the allocated

interventions adequately prevented?

Were patients blinded?

Were healthcare providers blinded?

Were data collectors blinded?

Were outcome assessors blinded?

Were data analysts blinded?

Definitely yes

Probably yes

Probably no

Definitely no

Was loss to follow-up (missing outcome data) infrequent?

Definitely yes

Probably yes

Probably no

Definitely no

Are reports of the study free of selective outcome reporting?

Definitely yes

Probably yes

Probably no

Definitely no

Was the study apparently free of other problems that could put it at a risk of bias?

Definitely yes

Probably yes

Probably no

Definitely no

Overall risk of bias

If applicable/necessary, per outcome measure

LOW

Some concerns

HIGH

Punt, 2014

Definitely no;

Reason: Study is a cluster-randomized trial in which the chosen intensive care

unit is the randomized element

No information

Reason: not described

Definitely no;

Reason: Study is open-label, no blinding

Probably yes;

Reason: Loss to follow-up was not assumed.

Probably yes;

Reason: All relevant outcomes seem to be reported. Study protocol is not accessible.

Probably yes;

Reason: Funding and conflicts of interest are not reported.

LOW (mortality)

Matchett,

2022

Probably yes;

Reason: Randomization sequence was computer-generated.

Definitely yes;

Reason: Opaque, sealed envelopes were used.

Definitely no;

Reason: Study is open-label, no blinding

Probably yes

Reason: Loss to follow-up was infrequent in intervention and control group. Adequate imputation methods (multiple imputation) were used

Definitely yes

Reason: All relevant outcomes were reported; Study protocol is available.

Probably yes;

Reason: One author receives compensation for serving on the Medical Advisory Board for the medical industry. This author contributed to the study design, enrolment, and manuscript drafting.

LOW (mortality)

HIGH (first pass success)

Reason: lack of blinding, other bias

Smischney, 2019

Definitely yes;

Reason: Computer-generated randomization and stratification by a

statistician independent of trial design were carried out in blocks of four patients based on sealed opaque randomization envelopes. The stratification was based on ICU location and hemodynamic compromise

Definitely yes;

Reason: Opaque,  sealed envelopes were used.

Probably yes;

Reason:

Healthcare providers were not blinded. Outcome assessors and statisticians were blinded. Patients were presumed to be not blinded.

Probably yes;

Reason: Loss to follow-up was infrequent in intervention and control group (≤6%).

Definitely yes

Reason: All relevant outcomes were reported; Study protocol is available.

Probably no;

Reason: First and second author have a provisional patent filed for the study drug (ketamine-propofol admixture).

However, the results suggest this did not skew the data.

LOW

(all outcomes)

Jabre, 2009

Probably yes;

Reason: computerised

random-number generator list provided by a statistician who was not involved in determination of patient eligibility, drug administration, or outcome assessment.

Probably yes;

Reason: the study drug was sealed in sequentially numbered, identical boxes containing the entire treatment for each patient.

Probably no;

Reason: emergency physician (who had  no

influence on the management of the patients) enrolling

patients was aware of study group assignment. Nurses and intensivists (caregivers) in the intensive care unit were blinded. Patients were presumed to be not blinded.

Probably no;

Reason: loss to follow-up was high (≥28%), yet similar in both groups.

Definitely yes

Reason: All relevant outcomes were reported; Study protocol is available.

Definitely no;

Reason: no other problems noted.

Some concerns

LOW (mortality)

Unclear (all other outcomes)

Reason: lack of blinding

Srivilaithon, 2023

Probably yes;

Reason: computer-generated randomization table with a block size of four by a statistician who was not involved in other study procedures

Definitely yes;

Reason: The drug allocation sequence was kept inaccessible to the research team throughout the study period. Patient assignments were placed into sequentially numbered sealed opaque envelopes

Probably no;

Reason: the study is single-blind, however it is not mentioned which roles are masked

Definitely yes;

Reason: no loss to follow-up

Definitely yes

Reason: All relevant outcomes were reported; Study protocol is available.

Definitely no;

Reason: no other problems noted.

Some concerns

LOW (mortality)

Unclear (all other outcomes)

Reason: lack of blinding

Tekwani, 2010

Probably yes;

Reason: computer-generated blocks of 10 randomization

Probably no;

Reason: not described

Probably yes;

Reason: study investigators and patients were blinded.

Probably yes;

Reason: Loss to follow-up was infrequent in intervention and control group (<3.5%) and handled adequately.

Definitely yes

Reason: All relevant outcomes were reported; Study protocol is available.

Definitely no;

Reason: no other problems noted.

LOW (all outcomes)

Ali, 2021

Probably yes;

Reason: computer-generated sequence of codes which was performed by a statistician

Probably yes;

Reason: The codes were inserted in opaque envelopes

which contained the instructions for preparation of study

drugs

Probably yes;

Reason: study was double-blinded, presumably subjects and investigators.

Probably yes;

Reason: Loss to follow-up was infrequent in intervention and control group (<17.5%)

Definitely yes

Reason: All relevant outcomes were reported; Study protocol is available.

Definitely no;

Reason: no other problems noted.

LOW (all outcomes)

Reference

Reason for exclusion

Hohl, C. M. and Kelly-Smith, C. H. and Yeung, T. C. and Sweet, D. D. and Doyle-Waters, M. M. and Schulzer, M. (2010)

Only partly conform current PICO

Albert, S. G. and Ariyan, S. and Rather, A. (2011)

Only partly conform current PICO

Cohen, L. and Athaide, V. and Wickham, M. E. and Doyle-Waters, M. M. and Rose, N. G. W. and Hohl, C. M. (2015)

Only partly conform current PICO

Sharda, Saurabh C. and Bhatia, Mandip S. (2022)

Only partly conform current PICO

Chan, C. M. and Mitchell, A. L. and Shorr, A. F. (2012)

Only partly conform current PICO

Baekgaard, J. S. and Eskesen, T. G. and Sillesen, M. and Rasmussen, L. S. and Steinmetz, J. (2019)

Only partly conform current PICO

Bruder, E. A. and Ball, I. M. and Ridi, S. and Pickett, W. and Hohl, C. (2015)

Only partly conform current PICO

Arteaga Velásquez J, Rodríguez JJ, Higuita-Gutiérrez LF, and Montoya Vergara ME. (2022)

Only partly conform current PICO

Rabelo, N. N. and Rabelo, N. N. and Machado, F. S. and Joaquim, M. A. S. and Dias Junior, L. A. A. and Pereira, C. U. (2016)

wrong publication type (narrative review)

De La Grandville, B. and Arroyo, D. and Walder, B. (2012)

wrong publication type (narrative review)

Trentzsch, H, Münzberg, M, Luxen, J, Urban, B, and Prückner, S. (2014)

Wrong language

Lysakowski C, Suppan L, Czarnetzki C, Tassonyi E, and Tramèr MR. (2007)

wrong comparison (true vs modified RSI)

Gu, W. J. and Wang, F. and Tang, L. and Liu, J. C. (2015)

not the most recent SR

Cabrini, L. and Landoni, G. and Baiardo Radaelli, M. and Saleh, O. and Votta, C. D. and Fominskiy, E. and Putzu, A. and Snak de Souza, C. D. and Antonelli, M. and Bellomo, R. and Pelosi, P. and Zangrillo, A. (2018)

Not conform PICO

Edwin, S. B. and Walker, P. L. (2010)

Outdated SR, not conform PICO

Kotani, Y. and Piersanti, G. and Maiucci, G. and Fresilli, S. and Turi, S. and Montanaro, G. and Zangrillo, A. and Lee, T. C. and Landoni, G. (2023)

original RCTs are included

Hildreth, A. N. and Mejia, V. A. and Maxwell, R. A. and Smith, P. W. and Dart, B. W. and Barker, D. E. (2008)

wrong population, wrong outcomes

Kim, M. H. and Oh, A. Y. and Han, S. H. and Kim, J. H. and Hwang, J. W. and Jeon, Y. T. (2015)

wrong population (elective surgery)

Meyancı Köksal G, Erbabacan E, Tunalı Y, Karaören G, Vehid S, and Öz H. (2015)

wrong outcomes

Kamali, A. and Taghizadeh, M. and Esfandiar, M. and Akhtari, A. S. (2018)

wrong outcomes

Smischney, N. J. and Beach, M. L. and Loftus, R. W. and Dodds, T. M. and Koff, M. D. (2012)

wrong population (not emergency)

Jaffrelot, M. and Jendrin, J. and Floch, Y. and Lockey, D. and Jabre, P. and Vergne, M. and Lapostolle, F. and Galinski, M. and Adnet, F. (2007)

wrong outcomes